Display driver nvlddmkm stopped responding, but has succesfully recovered

[viol37]

Hi Guys,

I recently upgraded both the motherboard and my video card and have been experiencing ongoing issues of the same symptoms with 3 of my most played games, which are BioShock, Counter-Strike Source and Crysis.

Issue will occur at random, which results in game play pausing for a few seconds then screen goes black.

Only way to get out of it is to perform a CTRL+ALT+DEL then terminate affected game via Task Manager, in which a Vista Bubble notification appears in the bottom right stating "Display driver nvlddmkm stopped responding, but has successfully recovered"

It seems that many people have been experiencing this issue for quite sometime (over a year) based on the searches I've done any have not been able to find any solutions.

My system specs are in my signature

Any thoughts and/or suggestions guys?

Viol37

PS:
I'm also running the latest version of ForceWare drivers, which is 182.08, which was released last week.

 

[NoStateofMind]

Originally posted by: viol37

My system specs are in my signature

?

 

[jaykishankrk]

Well generally speaking, this problem is directed more towards card's heat issues more than anything else.. i have come across this problem quite a few times now and i can assure you, every single time, my card's temps would have shot up by more than 85 deg.. so i would suggest you to monitor your video card's temperature and see if u can provide better cooling options for the same.. by the way my card is 8800GT which has a pathetic stock cooler..

 

[viol37]

Originally posted by: PC Surgeon

Originally posted by: viol37

My system specs are in my signature

?

Sorry about that mate,

I thought I had configured my signature correctly.

I have updated it.

 

[viol37]

Originally posted by: jaykishankrk
Well generally speaking, this problem is directed more towards card's heat issues more than anything else.. i have come across this problem quite a few times now and i can assure you, every single time, my card's temps would have shot up by more than 85 deg.. so i would suggest you to monitor your video card's temperature and see if u can provide better cooling options for the same.. by the way my card is 8800GT which has a pathetic stock cooler..

Hi mate,

I too had a feeling that it might have been related to a heat issue.

Yes my card does run hot (approx 60 idle, 85 load) but I have had other users also achieving those temps and have not had a problem.

Also I have seen that it is not only GTX295 users experiencing the issue where users out there with lower end nVidia card, which don't run as hot as mine still experiencing the issue.

People out there are annoyed as they do not know what is the cause of the issue so people as suggesting all sorts of solutions ranging from Motherboard, BIOS, Video card, RAM and PSU issues...

I don't know guys but I'm at a loss here

 

[imported_Redmist]

I get that error sometimes when I use my 9800gx2 card and play age of conan, everytime I had it, it was because my mobo temps were near the 65C mark and I would either get that error or my system would lock up.
I'm stuck using my 8800GT until I get a new case with better air flow.
given the florida heat, the 9800gx2 card, my case with only a exhaust fan in the rear, the hot nforce 560 chipset, and I just use the 8800GT until my new system arrives.

 

[LordGestle]

I build systems on the side and recently ran into this. Problem system below:

Q6600 (Stock)
ASUS P5Q-E
Ballistix PC6400 4GB (2*2GB)
Antec Quatro 1000 watt
WD 640GB Black.
EVGA 260 216 65nm.
Zalman 9700

What I tried:
1) Setting memory within bios manually. Even overvolted by +.1. Note: Overnight memtest ran fine, Windows Memory Checker no faults, OCCT no errors after 2 hours, Prime No errors over 8 hour span.
2) Disabled auto AI overclocker, manually set settings to stock, including DDR@1:1. No change. Also pulled cmos battery and cleared cmos for >4 hours. Same.
3) Swapped power supply (BFG LS-550), Memory (Corsair TwinX 4GB PC6400), and reverted -2 bios on the motherboard from current. Rinse and repeat above, no change.
4) Fresh install of Vista SP1, Vista SP2, and Windows 7. Windows 7 included drivers would not repeat issue. Updating anything >178 would repeat on Vista or Windows 7. I did not got any further back than 178 outside of the included windows 7 driver. All were 64 bit versions, but I did install 32bit Vista SP1 with no difference.
5) Replace with HD4830. Loops 3DMark06 and overnight Aquamark with no issue.
6) Place 260GTX in Abit IP35-E/Q6600/Same memory. No issue with fresh installations of Vista SP1. All stock.
7) Heat on GPU never exceeded >65-75. This was verified via EVGA prcesion on both systems, so temp was no different, but on the P45 chipset (P5Q-E) it would fail where as the P35 Chipset (Ip35-E) it would not display this issue. Both times the VGA card was ran stock. On the IP35-E the same VGA card coul reach and bench at SCC clocks with no problems. To add to the lack of heat the tested system was using a Antec Twelve Hundred (low and full speed) where as the IP35-E was in a case with only one intake, and one exhaust, both 120mm at 8-10 volts. Couldn't be any better on the Antec unless went water or after market gpu.

What was odd that I built an exact duplicate of this system 4-5 months ago with no issue. Only difference was the 260GTX was a PNY 260 GTX 192 and the CPU was an 8200. I have a spare P5Q-E and EVGA 260GTX on the way, but I'm guessing the motherboard is the culprit or at least a compatibility exists between the motherboard and the vga card.

In the past when I have seen video driver related stops/bsods they were all related to failing memory or incorrect bios settings. After running through the drill several times it's down to the motherboard as the culprit only due to I have not yet tested another 260gtx, but that same card works fine on the IP35-E (only PCI-E 1.0 x 16). Also What is odd is the HD4830 displays no such issue in the P5Q-E.

BTW, this can be easily reproduced using 3DMARK06 or Vantage. Usually fails to the desktop with nvlddmkm stopped responding within the first round of tests (<10 minutes).

Update: Also tested a 7600GT PCI-E in the same system. No issue. So either a combination of PCI-E 2.0 and flaky P5Q-E or EVGA 260 GTX. Will update next week when the parts arrive.

 

[viol37]

Originally posted by: LordGestle

BTW, this can be easily reproduced using 3DMARK06 or Vantage. Usually fails to the desktop with nvlddmkm stopped responding within the first round of tests (<10 minutes).

Thanks for the feedback guys.

I also have an update.

I have decided to return the card back to the place of purchase and will soon get a replacement.

In the meantime I am using my nVidia 8800GTS 512MB.

I have really given the card a workout by playing Crysis with all video settings on Very High with 2xAA enabled.

Game runs at very low FPS, which is expected but the purpose of the exercise is to try and replicate the issue with ""Display driver nvlddmkm stopped responding, but has successfully recovered"

Good news is that I was able to play Crysis for about an hour without the issue re-occuring.

I also ran 3DMark Vantage benchmark at maximum resolution, 2xAA, 16xAF with Extreme settings and ran it in a continuous loop for a few hours and my machine did not crash once.

And finally, also using the same ForceWare driver of 182.08, which I was using with my GTX295 and so far all is OK

Keep you guys posted...

 

[CoinOperatedBoy]

I am also experiencing this issue with an eVGA 8800GTS 320MB. One fix I've read is to uninstall the nVidia driver, search in Windows/System32 for all instances of nvlddmkm.sys and delete them, then reinstall the driver. This did NOT work for me, but others have reported success.

 

[rick475]

I had the same problem on VISTA 64 bits. My card wasn't running hot at all, with my speed fan set to 55% while gaming on rivatuner the card stays cool.

I had success with the driver sweeper. So far, 1 month without incident.

I believe this problem is related to a conflict of drivers with VISTA. I had the problem for the first time after updating my video drivers. Apparently VISTA always keeps some backup garbage and the drivers never uninstall completely....

Good luck !

 

[Mem]

Originally posted by: rick475
I had the same problem on VISTA 64 bits. My card wasn't running hot at all, with my speed fan set to 55% while gaming on rivatuner the card stays cool.

I had success with the driver sweeper. So far, 1 month without incident.

I believe this problem is related to a conflict of drivers with VISTA. I had the problem for the first time after updating my video drivers. Apparently VISTA always keeps some backup garbage and the drivers never uninstall completely....

Good luck !

Its not Vista,I had "Display driver nvlddmkm stopped responding, but has successfully recovered" in XP first anyway I solved the issue but also installed Vista after had got the same problem ,the problem turned out to be Factory OC video card (7800gt OC)not stable at factory OC settings,I underclocked my video card by 10Mhz or so and it was rock stable after on both OPERATING SYSTEMS.

Anyway since then I have upgraded to ATI 4870 card and installed new ATI drivers every month with NO issues on my Vista x64,even my factory 7800GT card was fine with any Nvidia drivers once its was underclocked by 10MHz(before I replaced it).

Conclusion is simple its not the OS but always something else ie overclock,heat,ram timings , video card etc...

 

[CoinOperatedBoy]

Originally posted by: Mem

Its not Vista,I had "Display driver nvlddmkm stopped responding, but has successfully recovered" in XP first anyway I solved the issue but also installed Vista after had got the same problem ,the problem turned out to be Factory OC video card (7800gt OC)not stable at factory OC settings,I underclocked my video card by 10Mhz or so and it was rock stable after on both OPERATING SYSTEMS.

Anyway since then I have upgraded to ATI 4870 card and installed new ATI drivers every month with NO issues on my Vista x64,even my factory 7800GT card was fine with any Nvidia drivers once its was underclocked by 10MHz(before I replaced it).

Conclusion is simple its not the OS but always something else ie overclock,heat,ram timings , video card etc...

I have also heard that it's usually a hardware issue and sometimes related to factory OC. For the record, my problem did not surface immediately after a driver update. I had been using the same driver for a few weeks before I first got the error. My video card is a factory OC "superclocked" unit.

Tonight I'm going to try Driver Cleaner/Sweeper. If that fails, I'll try to underclock the GPU and see if it helps. If it does, I assume I'd be eligible for an RMA since eVGA's cards are covered by limited lifetime warranties...

 

[CoinOperatedBoy]

Driver Sweeper didn't fix my issue. Nor did underclocking, although it seemed to limit the occurrence. Prior to underclocking, I was getting visual corruption/artifacts even when idle or performing regular OS tasks. Now those are gone, but I still can't run any 3D applications. I talked to eVGA support and they recommended an RMA.

 

[nerp]

I had this problem with my ATI 4850 and it was driving me nuts. The problem was fixed in the 9.1 catalyst drivers. I changed nothing with my setup, the card always ran cool. I was reading many forum threads where people insisted that it had nothing to do with software and was a hardware problem. They were wrong, apparently.

Seems the problem was common for ATI cards with dual display setups. The problem was triggered whenever I was watching full screen video. Wasn't consistent, somewhat random, but you could bet your bum that I'd be seeing the error a couple times a week without fail. A few times, as it recovered, the screen would blink repeatedly and sometimes result in an endless loop/blue screen. This was infuriating.

Now that ATI seems to have fixed the issue in their drivers, I haven't had a _single_ issue.

Now, factory OC cards seem to hgave this problem for both ATI and Nvidia cards. This might be the first place to look if the card is oc'd.